How Cryptocurrency Prices Are Analyzed

Forecasting any asset price is uncertain, and cryptocurrency is a hard case: markets trade 24 hours a day across every jurisdiction, the asset class is barely two decades old, and volatility runs far higher than in most traditional markets. No method described here is a crystal ball. Each is better understood as a lens: a way of turning some slice of the available information into a testable, probabilistic view.

The lenses have very different origins. Technical analysis predates crypto by more than a century: Charles Dow was formalizing trend analysis in the late 1800s, and candlestick charting is commonly traced back to eighteenth-century Japanese rice markets. On-chain analysis, by contrast, only became possible when Bitcoin launched in 2009, because a public blockchain publishes its full transaction history for anyone to audit. Sentiment and fundamental analysis both borrow from equity-market practice, but each had to be adapted to an asset class in which most tokens produce no cash flows.

Technical analysis studies past price and trading-volume data on the premise that prices move in trends and that recurring crowd behavior leaves recurring patterns. It is especially popular in crypto because markets are liquid, always open, and data-rich: charts update continuously, with no opening bell and no weekend gap.

The workhorse tools are moving averages, which smooth price into a single line. A simple moving average (SMA) is the arithmetic mean of the last N closing prices: if a coin closed at 100, 102, 104, 103, and 106 over five days, its 5-day SMA is 103. An exponential moving average (EMA) weights recent closes more heavily, so it reacts faster. Trend followers read the slope of these lines and watch crossovers: when a shorter average (often 50 days) rises above a longer one (often 200 days), the event is called a golden cross and is read as a bullish trend change, while the reverse is a death cross. Because averages are built from past prices, both signals confirm a move only after much of it has already happened.

The relative strength index (RSI), introduced by J. Welles Wilder in 1978, compares the size of recent gains to recent losses over a lookback window, conventionally 14 periods, and maps the result onto a 0 to 100 scale. Readings above 70 are conventionally labeled overbought and readings below 30 oversold. The 30/70 convention is a heuristic from Wilder's original work, not a law: in a strong trend the RSI can sit above 70 for weeks while price keeps rising, which is why practitioners pair it with trend context rather than trading it mechanically.

MACD (moving average convergence divergence), developed by Gerald Appel in the late 1970s, subtracts a 26-period EMA from a 12-period EMA and compares the result with its own 9-period average; crossings and divergences between the two lines are read as momentum shifts. Support and resistance are price zones where buying or selling has repeatedly clustered; analysts watch whether a level holds, breaks, or flips roles after a break. Volume acts as the confirmation layer throughout: a breakout on heavy volume is treated as more meaningful than the same move on thin trade.

How much of this actually predicts anything is a long-running academic debate. The efficient-market hypothesis, formalized in Eugene Fama's 1970 review, argues that past prices already reflect available information, so chart patterns should carry no exploitable signal. Andrew Lo's adaptive markets hypothesis (2004) offers a middle position: markets are mostly efficient, but inefficiencies appear and fade as conditions and participants change. A separate empirical literature on momentum, going back to Jegadeesh and Titman's 1993 study of US stocks, has documented that recent winners tend to keep outperforming over intermediate horizons, and several later studies have reported similar effects in crypto markets, though estimates vary and documented effects can weaken once widely known. Practitioners generally settle in the middle: technical analysis as a framework for timing, risk placement, and discipline rather than a forecasting machine.

On-chain analysis is largely unique to crypto. Because most blockchains are public ledgers, analysts can measure network activity directly, transaction by transaction, rather than inferring it from disclosures. The raw ledger differs by design: Bitcoin uses a UTXO model (unspent transaction outputs, where every coin is a discrete lump with its own acquisition date and price), while Ethereum uses an account model (running balances, closer to a bank ledger). The UTXO design is what makes several Bitcoin-specific metrics possible, because each coin's cost basis and age can be tracked individually.

Frequently watched metric families include:

• Active addresses and transaction counts: proxies for real usage of the network, watched for divergence from price.
• Exchange netflow: coins flowing onto exchanges (often read as intent to sell) minus coins flowing off into self-custody (often read as intent to hold).
• Realized capitalization: values each coin at the price when it last moved instead of the current price, approximating the aggregate cost basis of all holders.
• MVRV: market value divided by realized value. Elevated readings mean the average holder sits on large unrealized gains, a condition historically associated with cycle tops; readings below 1 mean the average holder is underwater.
• NVT: network value divided by on-chain transaction volume, sometimes described as a price-to-earnings analogue for blockchains, where high readings flag valuations running ahead of usage.
• SOPR (spent output profit ratio): whether coins being moved are, on average, being sold at a profit (above 1) or at a loss (below 1).
• Long-term holder supply: the share of coins that have not moved for an extended period (commonly 155 days in published methodologies), read as a gauge of conviction.
• Miner metrics: for proof-of-work chains, hash rate and miner outflows, since miners are structural sellers who must cover operating costs.

The main caveat is that on-chain data is precise about transactions but fuzzy about people. Attributing addresses to exchanges, funds, or individuals relies on heuristic clustering: pattern matching and labeling that analytics firms refine continuously but can never fully verify. One person can control thousands of addresses, exchanges shuffle coins internally between their own wallets, and a large transfer can be custodial housekeeping rather than a trading decision. On-chain metrics are therefore best read as slow-moving structural evidence rather than trade signals, and different data providers can publish materially different values for the same metric.

Sentiment analysis measures the mood and positioning of market participants rather than price or ledger data. The most cited instrument is the crypto Fear & Greed Index, which compresses several inputs (volatility, momentum and volume, social media activity, Bitcoin dominance, and search trends, in roughly that order of weight in the widely quoted alternative.me version) into a single 0 to 100 daily reading.

Beyond composite indexes, analysts practice social listening, tracking post volume, engagement, and tone across social platforms, where sudden spikes in attention often accompany local extremes. They also watch derivatives positioning. Perpetual futures funding rates measure what leveraged longs periodically pay shorts (or the reverse) to keep positions open: strongly positive funding means longs are paying up, which is read as crowded bullish positioning. Open interest, the total value of outstanding derivative contracts, shows how much leverage is in the system and therefore how much fuel exists for liquidation cascades in either direction.

The standard way to use sentiment is contrarian: extreme fear tends to coincide with prices that have already fallen, and extreme greed with froth. The equally standard failure mode is that extremes persist. A market can sit in extreme greed for weeks during a strong uptrend and in extreme fear through a long decline, so mechanically fading every extreme would have been costly in many past episodes. Sentiment works best as context for other evidence, not as a standalone trigger.

Fundamental analysis asks what a network should be worth, independent of near-term price action. Its crypto-native core is tokenomics: the rules governing a token's supply. Analysts examine the maximum supply (Bitcoin's is capped at 21 million coins), the issuance schedule (Bitcoin's new issuance halves roughly every four years, while other protocols emit on their own curves), and any burn mechanisms that remove tokens from circulation, such as Ethereum's EIP-1559 upgrade (2021), which destroys a portion of every transaction fee and ties net supply growth to network usage.

Other pillars include developer activity (sustained, distributed contributions to public code repositories, read as a sign the protocol is being maintained and improved), total value locked (TVL) for DeFi protocols (the value of assets deposited in a protocol's contracts, a rough gauge of trust and usage tracked by aggregators such as DefiLlama), and network effects, since a settlement network becomes more useful as more people, integrations, and liquidity connect to it.

The unresolved problem is valuation itself. Equity analysts anchor on discounted cash flow: a stock is worth the present value of the cash it is expected to return. Most tokens pay no cash flows to holders, so that anchor is missing, and the proposed substitutes (monetary premiums, fee capture, staking yields, transaction-demand models) remain debated. This is why fundamental views in crypto tend to be directional judgments about adoption and scarcity rather than precise price targets.

Cryptocurrency prices also move with forces entirely outside crypto. Global liquidity matters: when central banks are easing and real yields are falling, speculative assets broadly benefit, and crypto has historically been among the most sensitive. Interest-rate expectations, dollar strength, and risk appetite in equity markets all spill over.

The strength of these links shifts over time. Bitcoin traded largely independently of equities for much of its early history, then moved much more closely with technology stocks during the 2022 tightening cycle, a shift noted in IMF analyses published around that time. Correlation regimes are unstable in both directions, which is why analysts describe crypto's macro sensitivity as conditional rather than fixed: it tends to be highest when leverage is elevated and institutional participation is large.

No single lens answers every question. The table below summarizes what each examines, where it earns its keep, and where it fails.

Experienced analysts rarely act on one signal. The organizing idea is confluence: a view earns conviction when independent lenses agree. A trend reversal flagged on the chart carries more weight if exchange outflows show coins moving into storage, sentiment sits near extreme fear, and the macro backdrop is loosening. Because each lens has different blind spots, agreement across them is less likely to be noise.

Two supporting disciplines matter as much as the signals themselves. Timeframe separation keeps each lens in its lane: on-chain and fundamental evidence speaks to months and years, technical structure to days and weeks, and funding or sentiment extremes to hours and days. Mixing horizons, such as citing a four-year supply schedule to justify a leveraged day trade, is a common source of error. Risk sizing accepts that any single view is closer to a coin flip than practitioners like to admit, so positions are sized to survive being wrong; the effort goes into the process precisely because no individual call is reliable.

One practical way to test any of these methods is to make the forecast explicit and score it. Writing a timestamped, direction-and-horizon crypto prediction and letting it settle against the market turns an opinion into a measurable track record, which is how both individual analysts and prediction markets separate skill from noise.

Several beliefs recur among newcomers and do not survive contact with the evidence:

• A chart pattern is a guarantee. Patterns describe tendencies at best; every documented signal fails frequently, which is why sizing for being wrong matters more than pattern recognition.
• Indicators predict the future. RSI, MACD, and moving averages are transformations of past prices. They summarize what has happened; any forward-looking information in them is statistical and modest.
• On-chain data shows what individuals are doing. It shows what addresses are doing. Mapping addresses to actors is heuristic, and large transfers are often custodial operations rather than trading decisions.
• Extreme sentiment is a timing signal. Extremes flag conditions, not turning points. Markets have historically stayed greedy or fearful far longer than contrarians expected.
• More indicators mean better analysis. Indicators built from the same price series are highly correlated with one another. Five oscillators agreeing is one signal, not five.

The constraints on all of this deserve their own section. Cryptocurrency remains extraordinarily volatile: double-digit daily moves in major assets and peak-to-trough drawdowns exceeding 70 percent have occurred repeatedly across past cycles. Volatility does not just make forecasting harder; it makes otherwise sound analysis unsurvivable when positions are sized too large.

Markets are also reflexive: analysis feeds back into the thing it analyzes. Widely watched levels attract orders because they are widely watched, and a popular on-chain signal loses power as more traders act on it. Crypto is additionally exposed to discontinuous shocks: a regulatory action, an exchange failure, or a protocol exploit can reprice an asset in minutes in ways no chart, ledger metric, or sentiment gauge anticipated. The 2022 collapses of Terra and FTX are recent reminders that the largest moves often come from outside every model.

Past performance is not indicative of future results. Every method on this page is a way to structure thinking under uncertainty, not to remove it. Nothing here is financial advice.

Short definitions for terms used throughout this guide:

• Moving average (MA): the average closing price over a fixed lookback window, drawn as a smooth line over price. Simple (SMA) weights all periods equally; exponential (EMA) weights recent periods more.
• Golden cross / death cross: a shorter moving average crossing above (golden) or below (death) a longer one, conventionally the 50-day and 200-day.
• RSI (relative strength index): a 0 to 100 momentum oscillator comparing recent gains to recent losses; above 70 is conventionally overbought, below 30 oversold.
• MACD: the gap between a 12-period and a 26-period EMA, compared against its own 9-period average, used to read momentum shifts.
• Realized cap: total network value where each coin is priced at the moment it last moved on-chain, approximating the aggregate cost basis of holders.
• MVRV: market capitalization divided by realized capitalization; a gauge of aggregate unrealized profit or loss.
• NVT: network value divided by on-chain transaction volume; high values suggest valuation running ahead of usage.
• SOPR: spent output profit ratio; whether coins being moved are realizing profits (above 1) or losses (below 1) on average.
• Funding rate: the periodic payment between longs and shorts in perpetual futures, read as a gauge of leveraged positioning.
• TVL (total value locked): the value of assets deposited in a DeFi protocol's smart contracts.

This guide draws on standard academic and industry sources. Key starting points:

• Fama, E. (1970). Efficient Capital Markets: A Review of Theory and Empirical Work. Journal of Finance.
• Lo, A. (2004). The Adaptive Markets Hypothesis. Journal of Portfolio Management.
• Jegadeesh, N. and Titman, S. (1993). Returns to Buying Winners and Selling Losers. Journal of Finance.
• Wilder, J. W. (1978). New Concepts in Technical Trading Systems (the book that introduced the RSI).
• Nakamoto, S. (2008). Bitcoin: A Peer-to-Peer Electronic Cash System (the Bitcoin whitepaper).
• alternative.me, Crypto Fear & Greed Index methodology.
• Glassnode Academy and Coin Metrics documentation, regarding on-chain metric definitions such as realized cap, MVRV, SOPR, and long-term holder supply.
• DefiLlama documentation, regarding total value locked (TVL) methodology.
• IMF analyses (2022), regarding the rising co-movement of crypto and equity markets.

This page is maintained by the BitPredict team as a neutral educational reference and is reviewed periodically for accuracy.